sign in sign up
<<
Home , Active site, Peptide bond, Trypsinogen, Zymogen

Proc. Natl. Acad. Sci. USA Vol. 73, No. 11, pp. 3825-3832, November 1976

Role of proteolytic in biological regulation (A Review) * (limited / activation/control mechanisms) HANS NEURATH AND KENNETH A. WALSH Department of Biochemistry, University of Washington, Seattle, Wash. 98195 Contributed by Hans Neurath, September 7, 1976

ABSTRACT Many enzymes, hormones, and other physio- biologically active and probably the first step in logically active proteins are synthesized as inactive precursors degradation (12). The specificity of limited proteolysis is best () that are subsequently converted to the active form understood in terms of the three-dimensional structure of a by the selective enzymatic cleavage (limited proteolysis) of protein and of the attacking because the bonds. The ultimate agency of activating enzymatic containing the susceptible pep- function is limited proteolysis, either in a single activation step region of the protein substrate or in a consecutive series (cascade). The specificity of each ac- tide bond must fit the of the attacking protease in tivation reaction is determined by the complementarity of the order for residues of the substrate to interact with zymogen substrate and the active site of the attacking protease. primary as well as secondary binding sites of the (13). The sequence of consecutive activation reactions is regulated In general, limited proteolysis is therefore directed toward by the specificity of each enzyme, whereas the degree of am- surface loops and random segments of polypeptide chains rather plification of the initial stimulus is determined by the efficiency sheets. of each activating step. than toward internal domains, helices, or pleated Zymogen activation produces a prompt and irreversible re- The activation of zymogens usually occurs by proteolytic sponse to a physiological stimulus, and is capable of initiating cleavage of a in a region that is amino terminal new physiological functions. Typical examples are the processes relative to the active site of the protein. This may be a conse- of blood coagulation, fibrinolysis, complement activation, quence of the process of protein , which proceeds hormone production, metamorphosis, rtilization, supra- in the direction from the amino to the carboxyl end. If it is as- molecular assembly, and . The zymogens of the pan- as creatic , in particular, have served as models for sumed that the protein assumes its correct tertiary structure detailed studies of the nature of the molecular changes that are regions of the polypeptide chain are synthesized, the zymogen involved in the dramatic increase in enzymatic activity that will be formed prior to the enzyme. Were the activation peptide ensues upon limited proteolysis of the zymogen. attached to the carboxyl end, would be synthesized before , fibrin before fibrinogen, or collagen before In recent years, it has become evident that many proteins are procollagen. By synthesizing an inactivating prefix before synthesized as inactive precursors or zymogens and that these synthesizing the active portion of the protein molecule, pre- are subsequently converted to physiologically active forms by mature physiological function is avoided. the selective enzymatic cleavage of peptide bonds. This process The position of zymogen activation in the overall scheme of is known as zymogen activation, a term which initially was physiological control processes is diagrammatically shown in applied to the activation of precursors of proteolytic enzymes Fig. 1. The term zymogen is being used herein to denote in such as trypsinogen, , or procarboxypep- general an inactive precursor that can be converted to an active tidase (1). It is now apparent that the same type of reaction is protein by the cleavage of one or more peptide bonds. This involved in a great variety of biological processes, such as blood process is essentially irreversible because, in common with many coagulation, fibrinolysis, complement reaction, hormone pro- other hydrolytic reactions, proteolysis is an exergonic reaction duction, development, differentiation, and supramolecular under normal physiological conditions and there are no simple assembly, all of which involve zymogen activation in one or biological mechanisms to repair a broken peptide bond. In this more steps (2-9). In the present article, we shall attempt to show respect, zymogen activation differs in kind from the freely that activation by limited proteolysis is indeed an important reversible mechanisms of allosteric transition or covalent control element which can initiate new physiological functions modification (14). Whereas the latter are suited to maintain or or regulate preexistent ones. modulate a steady state of intermediary metabolites, zymogen Virtually all zymogen activation reactions require the en- activation, by virtue of its operational irreversibility, can effect zyme-catalyzed cleavage of a unique peptide bond by "limited unidirectional changes in the cellular environment and can proteolysis." This term was first introduced by Linderstrom- induce new physiological functions. This type of initiation is Lang and Ottesen (10) to describe the restrictive peptide bond more rapid than that regulated by the selective transcription cleavage that induces the conversion of ovalbumin to a different of a genome and is triggered by signals that operate entirely on crystalline form, plakalbumin, under the influence of the the post-translational level. Typical zymogen activation reac- bacterial protease . Numerous examples of limited tions are summarized in Table 1. proteolysis have since been described and studied in detail, such In some of these processes, the zymogen is converted to the as the tryptic conversion of chymotrypsinogen to active protein in a single step, whereas in others the process (1), the release by subtilisin of the amino-terminal segments of involves consecutive steps or cascades (2) which serve to amplify (11), and the conversion of proinsulin to insulin small stimuli to major physiological responses. Many zymogen (5). Limited proteolysis is the last step in the synthesis of many activation reactions may have remained undetected thus far because the precursor becomes activated prior to isolation. * By invitation. From time to time, reviews on scientific and techno- Indeed, isolation procedures are usually designed for maximum logical matters of broad interest are published in the PROCEED- yield of active protein rather than of zymogen and thus may INGS. contravene the demonstration of a zymogen precursor. 3825 Downloaded by guest on September 27, 2021 3826 Biochemistry: Neurath and Walsh Proc. Natl. Acad. Sci. USA 73 (1976) AMINO ACID POOL

(b) (o) (b)

(C) (d) ZYMOGEN CTIV NACTIVE Limited PROTEIN I I Allosteric PROTEIN Proteolysis Transition or Covalent Modification FIG. 1. Schematic representation of major control mechanisms. (a) Transcription and translation regulate the rate of formation of the various proteins from the amino acid pool. (b) Other controls regulate the rate of degradation of the various proteins to their constitutent amino acids. (c) The activity of the protease that catalyzes zymogen activation may be in regulated by a series of consecutive reactions of limited proteolysis (Fig. 2). Activation of a zymogen is essentially irreversible in vivo. (d) Reversible conformational changes are responsive to effector concentrations or to the activities of specific group and .

CONSECUTIVE ZYMOGEN ACTIVATION trypsinogen by enterokinase is so specific that only a single bond REACTIONS out of 228 in trypsinogen is cleaved and no other protein has A series of consecutive zymogen activation reactions is shown yet been reported to be a substrate for enterokinase (16). Thus, diagrammatically in Fig. 2. X, Y, and Z are zymogens, each the site of generation of active trypsin is restricted to the con- having the potential of being converted to an active protein. fluence of these two secretory streams (Fig. 3). Active trypsin Conversion of the zymogen X to the protease Xa is triggered by in turn catalyzes the conversion of other pancreatic zymogens a specific physiological stimulus; in the ensuing cascade, the to their active forms, i.e., the chymotrypsinogens, proelastase, product of one reaction is a catalyst for the next. The sequence the procarboxypeptidases, and prophospholipase. This system of the events is determined by the specificity of each enzyme constitutes a two-stage cascade. and the degree of amplification of the initial stimulus is de- A more complex and extensive cascade system is found in the termined by the efficiency of each activating step. For instance, blood coagulation process, shown in Fig. 4. In fact, the term one molecule of Xa might produce 103 molecules of Ya, which cascade or waterfall was introduced by Macfarlane (17) and in turn produce 106 molecules of active protein. When the zy- by Davie and Ratnoff in 1964 (18) in connection with this series mogen is produced by one type and the activating protease of reactions. Five known proteolytic reactions occur along the by another, communication between the two cell types adds so-called intrinsic pathway, which is mediated entirely by another element to the control mechanism (15). For instance, components found in the plasma, and four proteolytic steps the activation of pancreatic trypsinogen, which originates in accompany the extrinsic pathway, which includes factors found the acinar cells of the , is triggered by enterokinase, in tissues (3). The intrinsic and the extrinsic pathways converge which is secreted from the brush border of the small intestine to produce , which in turn converts fibrinogen to fi- (16). It should be noted, parenthetically, that the activation of brin.

Table 1. Examples of zymogens that are converted to active proteins in response to various stimuli Programmed response in Normal physiological response Response to foreign stimulus development or repair Vasoactive products Digestion Development Angiotensinogen (61) Pepsinogen Prochitin synthetase (7) Prekallikrein* Prochymosin Prococoonase (30) Trypsinogen* Procollagenase (68) Hormones Chymotrypsinogen Proacrosin (31) Proinsulin Procarboxypeptidases Fibrinolysis Proglucagon (62) Prophospholipase (67) Plasminogen proactivator* Proparathyroid hormone Proelastase Plasminogen Enzymes Blood coagulation Toxin Protyrosinase (63, 64) Factors VII*, IX*, X* Promellitin (69) Prephenoloxidase (65) Prothrombin* Self-assembly Prorenin (66) Factor XIII Procollagen Fibrinogen Phage head proteins Complement Processing of zymogen precursors Properdin precursor (34-37) Factors C3, Cs * Except where indicated by a specific reference, references to these zymogens are found in any of three general sources: Proteolytic Enzymes, Methods in Enzymology, Vol. 19 (77); Peptide Bond , The Enzymes, Vol. 3 (78); and Proteases and Biological Control, Cold Spring Harbor Conferences on Cell Proliferation, Vol. 2 (79). * An intermediate component in a series of consecutive zymogen activations. Downloaded by guest on September 27, 2021 Proc. Natl. Acad. Sci. USA 73 (1976) Regulation by proteases (Review) 3827

TRIGGER *

X X XII XIao Y Ya 7 Active Z Protein XI X10 tissue factor FIG. 2. Schematic representation of consecutive zymogen acti- 00+2 vation reactions (cascade). Activation of the zymogens X, Y, and Z leads to the formation of the proteases Xa and Y. and the active IX IXa VII a VII protein, in response to a trigger which initiates the conversion of X to Xa. pnospholipi phospholipid X Xo X In addition to the multiple activation steps shown in Fig. 4, /Ca+2 processes such as blood coagulation appear to be regulated by phospholipid specific protein inhibitors (19), which terminate the activation prothrombin thrombin reactions once the product has been generated and restrict the site of blood coagulation to the area of the injury. In this case, a pulse of protease is generated only in the brief interval be- f ibrinogen f brin tween zymogen activation and the subsequent formation of an FIG. 4. An abbreviated representation of the cascade of consec- inactive enzyme-inhibitor complex, as represented schemati- utive zymogen activations involved in blood coagulation (3). The in- cally in Fig. 5. This general scheme illustrates that the physio- trinsic pathway (left) and the extrinsic pathway (right) converge in logical signal is transduced to a chemical event by conversion activating . The chemical nature of the initial stimulus (*) of zymogen X to protease X., which in turn converts Y to Y, and is not completely understood. is then inactivated by the inhibitor Ix. In turn, Y. exists long enough to activate Z and is then inhibited by ly. Thus at each dicated schematically in Fig. 6, zymogens are inactive because step, the duration and breadth of the cascade is controlled to the activation peptide confers an altered geometry on the some extent by the concentration of inhibitors, whereas the molecule. The structure of the activation peptide is unrelated degree of amplification is determined by the number of steps to that of the active site of the enzyme. On the other hand, between Xa and Z. The ultimate change in microenvironment macromolecular inhibitors contain regions that are comple- is determined by the nature of the active protein that is finally mentary in shape to the active site of the enzyme and some have expressed. There are only a few examples of cascading systems the characteristics of a pseudosubstrate (22, 23). In fact, only of this kind, but certainly now that the principle is established a few residues of the inhibitor molecule interact with the active many more will be found. For instance, premature activation site of the enzyme, whereas the bulk of the inhibitor serves of the pancreatic enzymes shown in Fig. 3 is inhibited at the mainly the function of a supporting structure. This is true of the level of protease Y (trypsin) by pancreatic trypsin inhibitors. pancreatic and soybean trypsin inhibitors, as well as of the The various intermediate proteases of the blood coagulation inhibitor from potatoes that has been recently system are inactivated by specific inhibitors, such as anti- investigated in our laboratory (24). thrombin III and a2-macroglobulin, whereas fibrinolysis (i.e., Cascades of zymogen activation reactions do not necessarily the dissolution of the fibrin clot) and the complement reaction operate in isolation but may influence one another in the initial each involve different sets of proteolytic activation reactions stages, by positive or negative feedback regulation, thus adding and different inhibitors such as al-antitrypsin, a2-macroglob- another element of control. For instance, three plasma activa- ulin, Cl-inactivator, and others (20, 21). tion systems, the coagulation system, the fibrinolysis system, It seems important at this juncture to emphasize the funda- and the system, interact with one another at key mental difference between zymogens on the one hand and points as shown diagrammatically in Fig. 7. The activation of enzyme-inhibitor complexes on the other. Both appear inert Hageman factor (factor XII) is enhanced by kallikrein, which by activity assays and both can be converted to the active en- in turn is generated from prekallikrein under the influence of zyme under the appropriate circumstances. However, as in- activated Hageman factor (3, 25, 26). Kallikrein in turn cata-

Secretion

,, - - NI N Intestinal / Brush Border Enteroki no se \ Cell v A 5ect-e,°o-%%-w Trypsinogen Trypsi n Chymotrypsinogens Chymotrypsi ns Poncreati cr\oeti Proelostose Acinor Cell Elastose Procorboxypeptidoses Corboxypeptidoses Prophosphol pose Phospholipose FIG. 3. Activation of zymogens in the involves a two-step cascade. Pancreatic trypsinogen is converted to trypsin by enterokinase. Trypsin in turn converts the other zymogens to active enzymes. Downloaded by guest on September 27, 2021 3828 Biochemistry: Neurath and Walsh Proc. Natl. Acad. Sci. USA 73 (1976)

ZYMOGEN PROTEIN PROTEIN D Protease ,---non Signal Transductio n X ~~~XIx -X I Iz nor- nonactive- active b active Protease Iy Amplification y - -~ Ya Iy (b)

Expression Z ACTIVE PROTEIN ZYMOGEN ENZYME E-I COMPLEX FIG. 5. Generalized scheme of control processes regulated by FIG. 6. Schematic illustration of the control of enzyme activity zymogens, proteases, and their inhibitors. Zymogens (X, Y, and Z) by (a) zymogen activation, (b) formation of an enzyme-inhibitor are activated sequentially in processes such as blood coagulation or complex, and (c) dissociation of that complex. -non represents the complement activation. Protease intermediates (Xa, Ya) may be in- activation peptide. activated by specific inhibitors (I., Iy) to limit their action. A physi- ological signal (*) initiates the cascade by converting zymogen X to protease Xa. proteolysis (34). While the precise significance of these events is currently under investigation, it seems certain that limited lyzes the formation of kinins (e.g., bradykinin) from their re- proteolysis, analogous to zymogen activation, transforms one spective precursors, the (27). Activated Hageman precursor to another and that the mechanisms of conversion of factor is also believed to enhance the conversion of plasminoger zymogens to enzymes may serve as models for the transfor- proactivator to the activated form, though this process may mation of prezymogen to zymogen. involve a number of as yet unrecognized intermediates (26). Another focus of interaction of these systems is the degradation EVOLUTION OF ZYMOGEN ACTIVATION of fibrin by . Some of these reactions can also be in- If limited proteolysis is indeed a significant physiological control hibited by intermediates of the activation reactions as well as mechanism, how did it come about during biological evolution? by specific plasma protease inhibitors (28, 29). Did the primordial cells contain zymogens that subsequently At present, descriptions of cascade processes deal primarily became enzymes, or did they contain enzymes that later be- with plasma proteases, but fragmentary evidence is consistent came inactivated by extension of the polypeptide chains so as with the occurrence of such complex processes elsewhere. For to form zymogens? The available evidence seems to-favor, but example, it is not yet clear whether the proteases that activate not prove, the latter alternative, which is also consistent with procollagen (8), proinsulin (5), prococoonase (30), and pro- the empirical observation that relatively few zymogens have (31) are directly triggered by specific enzymes or been found in prokaryotes or in less highly differentiated eu- whether consecutive protease-mediated steps control these karyotic organisms. transformations. The sequence of discoveries in this field is the It is a well-recognized fact that enzymes such as the mam- reverse of the physiological sequence of events because the malian serine proteases are a homologous set of proteins which starting point for investigation is usually the final product. operate by analogous catalytic mechanisms (38, 39). During Subsequent discovery of a precursor of this product leads to evolution, those amino acid residues have been preserved that inquiries about a catalyst for this transformation; this in turn are essential for function, for the maintenance of conformation, leads to questions about the origin of this catalyst, and so on. or for both. In enzymes, these regions include, in particular, Ultimately one must identify for each process the mechanism sequences around the active site, and in zymogens they might that triggers the first step (compare Table 1). be expected to include regions around the site of activation. In Experimental investigations have usually focused on the the case of the seine proteases, the catalytic apparatus includes physiologically active protein generated by zymogen activation the so-called "charge-relay" system (40) formed by interaction (e.g., an enzyme, hormone, or self-assembling protein), whereas of amino acid residues which are widely separated from each the peptide fragment has been generally ignored. However, other in the linear sequence of the molecule. These include, in recent studies of the activation of complement factor C3 indi- particular, 102, 57, and serine 195. The cate (32) that a physiologically active peptide, anaphylatoxin alpha amino group of isoleucine 16 forms a salt linkage with (Ca) is released from Cs together with an active protease (Cab). aspartic acid 194. (The numbering system is that of bovine Analogo'usly, complement factor C5 and prothrombin each chymotrypsinogen A.) An abbreviated presentation of such yield upon activation both an active protease and a pharma- patterns of homology among serine proteases is given in Fig. cologically active peptide (15, 32,33). It may be suggested that 8. The enzymes include three plasma proteases and their other activation processes also produce dual physiological homology with the pancreatic serine proteases suggests that products in response to a single stimulus and that such reactions both families of enzymes have diverged from a common an- serve to express multiple messages in a concerted manner. cestor (41, 42). Activation reactions involving limited proteolysis have re- Comparison of the zymogens of these proteases, on first sight, ceived added significance from the recent observations that indicates that they are so different in size, composition, and secretory proteins, such as the pancreatic zymogens, contain sequence as to preclude a common ancestral relationship. For an additional polypeptide extension on the amino side of the instance, in the case of bovine trypsinogen, the activation normal activation peptide (347). Semantically, these proteins peptide is composed of six residues, whereas in the other ex- are denoted by the prefix pre-, whereas zymogens are usually treme, the fragment released during the activation of pro- denoted by the prefix pro- or the suffix -ogen (e.g., prepro- thrombin is larger than the enzyme thrombin itself. However, parathyroid hormone, pretrypsinogen). These "prezymogens" upon closer inspection, at least three classes of activation pep- are thought to be the first products of messenger translation and tides seem evident. As shown in Fig. 9, trypsinogen activation evidence indicates that after passage of the nascent proteins are homologous and related to the activation peptide through the membrane of the endoplasmic reticulum the released during the generation of the moth enzyme cocoonase peptide extension is severed from the zymogen by limited from its precursor. Analogously, the activation peptides of Downloaded by guest on September 27, 2021 Proc. Natl. Acad. Sci. USA 73 (1976) Regulation by proteases (Review) 3829

Factor XII

Factor XIIo

Plasminogen Plasminogen pro-activator activator Factor XI XIO Prekallikrein Kallil'rein

Factor IX IXa Kininogen Kinin Plasminogen Plasrmin /o s Factor X Xa

l Prothrombin Thrombin

Fi brinogen Fibrin

Fragments FIG. 7. Interactions among some plasma zymogens and proteases in the fibrinolytic, coagulation, and kinin systems (7, 25-27). For details, see the text.

various chymotrypsinogens (A, B, and C) and proelastase are lated, one would conclude that the homologous enzymes arose homologous, although they appear to be quite different from first and that the prefixes were subsequently added as inde- those of the trypsinogen group. A third structural class of acti- pendent events following duplication and divergence of vation peptides comprises certain blood coagulation proteases, functions (45). However, one cannot rule out the possibility that particularly those of prothrombin, factor X, and factor IX. The the activation peptides of the serine proteases represent distant structural relationship among these zymogens is shown sche- homologous sequences which are the products of a rapidly matically in Fig. 10. Whereas prothrombin and factor IX are mutating gene, as for instance in the case of the fibrinopeptides. synthesized as single polypeptide chains and only subsequently de Haen et al. (46) have in fact argued for such a homologous during activation are converted into two-chain structures, factor relationship among these various zymogens. X appears to consist even initially of two chains (7). The The activation peptides of the three blood coagulation zy- amino-terminal sequences of the single chains of prothrombin mogens are also of interest because they illustrate that the re- and factor IX and of the light chain of factor X show evidence moval of the activation peptide during zymogen activation may of homology, but in the case of prothrombin and factor X no have additional consequences besides the generation of an ac- such similarity is evident in the sequences immediately pre- tive site. It has been recently demonstrated by several workers ceding the peptide bond that is cleaved during the activation that the amino-terminal region of prothrombin, and probably reaction (43, 44). If in fact the activation peptides of the three also of factors X and IX, contains an unusual amino acid, -y- groups of serine proteases, i.e., the trypsinogens, the chymo- carboxyglutamic acid, which appears to form an effective site trypsinogens, and the blood coagulation proteases, are unre- for chelating calcium (43, 47-49). Formation of this amino

PANCREATIC PROTEASES 16 57 102 195 Bovine chymotrypsin A Ile-Val-Asn-Gly-----Ala-Ala-His-Cys-----Asn-Asp-Ile -----Gly-Asp-Ser-Gly- Bovine trypsin Ile-Val-Gly-Gly- -Ala-Ala-His-Cys- -Asn-Asp-Ile- -Gly-Asp-Ser-Gly- Porcine trypsin Ile-Val-Gly-Gly- -Ala-Ala-His-Cys- -Asn-Asp-Ile- -Gly-Asp-Ser-Gly- Dogfish trypsin Ile-Val-Gly-Gly- -Ala-Ala-His-Cys- -Asn-Asp-Ile- -Gly-Asp-Ser-Gly- Porcine Val-Val-Gly-Gly- -Ala-Ala-His-Cys- -Tyr-Asp-Ile- -Gly-Asp-Ser-Gly-

PLASMA PROTEASES (Bovine) Thrombi n Ile-Val-Glu-Gly- -Ala-Ala-His-Cys- -Arj-Asp-Ile- -Gly-Asp-Ser-Gly- Factor Xa Ile-Val-Gly-Gly- -Ala-Ala-His-Cys- -Phe-Asp-Ile- -Gly-Asp-Ser-Gly-

Factor IXa Val-Val-Gly-Gly- n.d. ------n.d.---- -Gly-Asp-Ser-Gly-

Plasmin (human) Val-Val-Gly-Gly- -Ala-Ala-His-Cys- ---- n.d.---- -Gly-Asp-Ser-Gly-

OTHER PROTEASES

Cocoonase Ile-Val-Gly-Gly- n.d.------n.d.---- -Gly-Asp-Ser-Gly- FIG. 8. Similarities in amino acid sequences immediately adjacent to components of the active sites of various serine proteases (43, 44, 48, 70-75). The residue numbers are those of chymotrypsinogen A. Underlined residues differ from the majority shown for a particular position in the sequence. n.d. = not determined. Downloaded by guest on September 27, 2021 3830 Biochemistry: Neurath and Walsh Proc. Natl. Acad. Sci. USA 73 (1976)

GROUP I Bovine trypsi nogen Val -Asp-Asp-Asp-Asp-Lys Ile- Dogfish trypsinogen Al a-Pro-Asp-Asp-Asp-Asp-Lys Ile- Prococoona se (5 residues) ---Arg-Thr-Gln-Asp-Asp-Gly-Gly-Lys Ile-

GROUP II Bovine chymotrypsinogen A (7 res)---Pro-Val-Leu-Ser-Gly-Leu-Ser-Arg I 1e- Bovine chymotrypsinogen B (7 res)---Pro-Val-Leu-Ser-Gly-Leu-Ala-Arg Ile- Bovine chymotrypsinogen C (5 res)---Phe-Gln-Pro-Asn-Leu-Ser-Ala-Arg IVal- Lungfish proelastase A (3 res)---Pro-Ser-Tyr-Pro-Pro-Thr-Ala-Arg IVal-

GROUP III Bovine prothrombi n Ala-Asn-Lys-Gly-Phe-Leu-Gla-Gla------(307 res) ------Phe-Glu-Ser-Tyr-Ile-Glu-Gly-Arg I le- Bovine Factor X Ala-Asn-Ser- - -Phe-Leu-Gla-Gla--(133 res) (43 res)--Glu-Asp-Gly-Ser-Gln-Va1-Vai-Arg Ile- Bovine Factor IX. Tyr-Asn-Ser-Gly-Lys-Leu-Glx-Glx---- (ca. 147 res) ------Val - Bovine plasminogen Glu-Pro-Leu-Asp-Asp-Tyr-Val-Asn----(ca. 550 res)------Arg Val - FIG. 9. Homologous amino-terminal sequences of three groups of zymogens of serine proteases (30, 41, 43, 44, 46, 48, 70, 76). The vertical arrow denotes the site of limited proteolysis during zymogen activation. Gla indicates y-carboxyglutamic acid. acid is a prerequisite for the activation of prothrombin by factor provides the determinants of its three-dimensional structure X, and its absence in certain patients is associated with disorders and that chemical alterations of the backbone or the side chains of the blood coagulation process (49). The activation of pro- could disrupt the balance of forces that dictate a particular thrombin by factor Xa is accelerated by calcium ions, as is the conformation. A change in conformation could induce the de activation of factor X by factor IXa. Both activation reactions novo assembly of the catalytic apparatus, the generation of the require, in addition, phospholipid and other coagulation pro- substrate-, or the removal of an obstruction from teins (factors V and VIII, respectively). Whereas activated both (see Fig. 11). In the best-studied cases, i.e., the x-ray factor X (Xa) stays bound to the phospholipid micelle, thrombin crystallographic comparison of the structures of chymo- does not, because the calcium-binding site is released during trypsinogen and chymotrypsin, the, observed differences were the activation of prothrombin, but the binding site of factor X so subtle that it was not possible to decide unambiguously which remains attached to this zymogen even after activation (3). structural changes were crucial for zymogen activation and which were incidental (50). The recent discovery that many MECHANISM OF ZYMOGEN ACTIVATION zymogens possess weak but intrinsic enzymatic activity has It is of foremost interest to inquire about the mechanism by made it possible to examine the activation process by kinetic which the cleavage of a single polypeptide bond in a zymogen generates enzymatic function. Does limited proteolysis merely Zymogen remove an obstruction from the active site, or does it induce a conformational change that generates an active site? It is a 0] well-accepted dogma that the amino acid sequence of a protein

Site of act va t on

PROTHROMBIN 7CES,. (M, 70,000)

FACTOR X xII / JIIL.:ZiZZ :Z{..:i: (Mr 53,000)

FACTOR IX Substrate Catalytic (M, 55,000) Binding Site Apparatus FIG. 11. Schematic illustration of three possible modes of zy- TRYPSINOGEN mogen activation. Pathway A: the conversion of zymogen to enzyme (Mr 23,000) induces a conformational change that improves the substrate binding site. Pathway B: the binding site preexists in the zymogen and zy- FIG. 10. Comparison of the lengths of the polypeptide chains mogen activation induces the formation of an effective catalytic site. of the zymogen forms of four homologous serine proteases. The Pathway C: the activation peptide (hatched) occludes in the zymogen stippled segments denote the homologous enzymes and the dots the the entire active site, which becomes exposed only after removal or reactive . Mr = molecular weight. dislocation of this peptide segment. Downloaded by guest on September 27, 2021 Proc. Natl. Acad. Sci. USA 73 (1976) Regulation by proteases (Review) 3831 and spectral analysis of the zymogen before activation and of binding and aligning the substrate in an optimally productive the enzyme afterwards (51-55). These studies, to be described mode. briefly in the following paragraph, indicate that the catalytic apparatus of trypsin, chymotrypsin, and probably other zy- CONCLUSIONS mogens of serine proteases, is largely preexistent in the zymogen While the activations of pancreatic zymogens such as trypsin- forms and that during activation the effectiveness of the binding ogen and chymotrypsinogen are the best understood examples site is improved more than 1000-fold. The evidence can be of the induction of biological activity by limited proteolysis, it briefly summarized as follows. is important to emphasize that they represent only examples Several zymogens, like their respective enzymes, react stoi- of a great variety of physiological functions that are induced chiometrically with active-site-directed inhibitors such as di- by limited peptide of precursor proteins. The isopropylphosphofluoridate to form stable covalent compounds product of activation may be an enzyme, a hormone, a phar- at a which is 104 to 105 times slower than that of macologically active peptide, or a structural component of enzyme (52). Certain other pseudo-substrates react not only tissues, but the ultimate chemical event is limited proteolysis with the enzyme but also with the parent zymogen. In both in every instance. Proteases can thus generate functions but they cases stable acyl-intermediates can be isolated at low pH and can also destroy them, and in this sense it is useful to think of then deacylated at higher pH (54, 55). The second-order rate limited proteolysis as a control element that can turn other re- constants of are 104 to 105 times lower for the zymo- actions on and off by generating or destroying their catalysts. gen than for the enzyme, but the deacylation rates differ only The on and off reactions are controlled by different switches, by factors ranging from 2 to 70. The pseudo-substrate methane so to speak, because proteolysis is essentially an irreversible sulfonyl fluoride, which has a low affinity for the enzyme, process and proteases are not endowed with repair functions. presumably because it is not bound to the substrate-binding Zymogens are poised to respond to signals, to amplify them, and pocket, also has a low affinity for the zymogen but the sec- to respond to them irreversibly. Examples of such control ond-order rate constant differs not by 104 or 105as in the case processes are being increasingly observed and should be further of diisopropylphosphofluoridate, but by only about 50-fold (56). thought of in the exploration of the control of and Finally, a competitive inhibitor (p-aminobenzamidine) is bound other physiological phenomena, including processes of devel- by trypsinogen 103 to 104 times less firmly than by trypsin, opment and differentiation. presumably because the substrate-binding pocket is not fully developed prior to activation (54). All of these observations agree with the idea that in these zymogens the substrate-binding This work has been supported in part by grants from the National pocket is distorted but that the catalytic apparatus is largely Institutes of Health (GM15731) and the American Cancer Society preexistent. (BC-91R). It should be recognized, however, that the effectiveness of enzymes is determined by the geometries of both the enzyme 1. Neurath, H. (1957) Adv. Protein Chem. 12,320-386. and the to 2. Neurath, H. (1975) in Proteases and Biological Control, eds. substrate and that these have be compatible to form Reich, E., Rifkins, D. B. & Shaw, E. (Cold Spring Harbor Labo- the productive transition state during [in the case of ratory, Cold Spring Harbor, N.Y.), Vol. 2, pp. 51-64. serine proteases this appears to be a tetrahedral intermediate 3. Davie, E. W. & Fujikawa, K. (1975) Annu. Rev. Biochem. 44, (57, 58)]. With ideal substrates, the geometries of enzyme and 799-829. substrate are suitable to meet these requirements, whereas with 4. Muller-Eberhard, H. J. (1975) Annu. Rev. Biochem. 44, 697- poor substrates and pseudosubstrates the complex is less pro- 724. ductive because of misalignment of susceptible bonds relative 5. Steiner, D. F., Kemmler, W., Tager, H. S., Rubenstein, A. H., to the idealized state (M. A. Kerr, K. A. Walsh, and H. Neurath, Lernmark, A. & Zuhlke, H. (1975) in Proteases and Biological manuscript in preparation). Zymogens may be relatively poor Control, eds. Reich, E., Rifkin, D. B. & Shaw, E. (Cold Spring catalysts for a certain reaction because the geometry of the Harbor Laboratory, Cold Spring Harbor, N.Y.), Vol. 2, pp. active site is 531-549. unfavorable to bind substrates in a productive mode 6. Zanefeld, L. J. D., Polakoski, K. L. & Schumacher, G. F. B. (1975) and to form the idealized transition state. Chymotrypsinogen in Proteases and Biological Control, eds. Reich, E., Rifkin, D. and trypsinogen are inferior catalysts for all known substrates B. & Shaw, E. (Cold Spring Harbor Laboratory, Cold Spring but activation by limited proteolysis changes the conformation Harbor, N.Y.), Vol. 2, pp. 683-706. of these zymogens, improves the binding of specific substrates, 7. Cabib, E. & Farkas, V. (1971) Proc. Natl. Acad. Sci. USA 68, and also allows the transition state to be reached. 2052-2056. In a few cases, however, such conformational transitions 8. Bornstein, P. (1974) J. Supramol. Struct. 2, 108-120. occur seemingly without peptide bond cleavage. For instance, 9. Laemmli, U. K. (1975) in Proteases and Biological Control, eds. the conversion of human plasminogen to plasmin is induced by Reich, E., Rifkin, D. B. & Shaw, E. (Cold Spring Harbor Labo- and as a first the formation of a of ratory, Cold Spring Harbor, N.Y.), Vol. 2, pp. 661-676. streptokinase, step complex 10. Linderstrom-Lang, K. U. & Ottesen, M. (1949) C. R. Trav. Lab. streptokinase and plasminogen alone suffices to induce enzy- Carlsberg 26, 403-442. matic activity without peptide bond cleavage (59,60). Similarly, 11. Richards, F. M. & Vithayathil, P. J. (1959) J. Biol. Chem. 234, the activation of factor XII (Hageman factor) and of comple- 1459-1465. ment factor C1 does not appear to involve peptide bond 12. Schimke, R. T. & Doyle, D. (1970) Annu. Rev. Biochem. 39, cleavage (4,26). In the case of factor X, peptide bond cleavage 929-976. is required for activation but the differences in enzymatic ac- 13. Ottesen, M. (1967) Annu. Rev. Biochem. 36,55-76. tivity toward pseudosubstrates between zymogen and active 14. Stadtman, E. R. (1970) in The Enzymes, ed. Boyer, P. D. (Aca- enzyme are not nearly as great as in the case of the model sys- demic Press, New York), 3rd ed., Vol. 1, pp. 397-459. tem, chymotrypsinogen and chymotrypsin (M. A. Kerr, K. A. 15. Walsh, K. A. (1975) in Proteases and Biological Control, eds. and H. Reich, E., Rifkin, D. B. & Shaw, E. (Cold Spring Harbor Labo- Walsh, Neurath, manuscript in preparation). It thus ratory, Cold Spring Harbor, N.Y.), Vol. 2, pp. 1-11. appears that the various zymogens and their enzymes differ 16. Maroux, S., Baratti, J. & Desnuelle, P. (1971)J. Biol. Chem. 246, from each in their relative catalytic efficiency, depending upon 5031-5039. the magnitude of the conformational changes required for 17. Macfarlane, R. G. (1964) Nature 202,498-499. Downloaded by guest on September 27, 2021 3832 Biochemistry: Neurath and Walsh Proc. Natl. Acad. Sci. USA 73 (1976) 18. Davie, E. W. & Ratnoff, 0. D. (1964) Science 145, 1310- 46. de Haen, C., Neurath, H. & Teller, D. C. (1975) J. Mol. Biol. 92, 1312. 225-259. 19. Heimburger, N. (1974) in Bayer-Symposium V, Proteinase In- 47. Stenflo, J. (1974) J. Biol. Chem. 249,5527-5535. hibitors, eds. Fritz, H., Tschesche, H., Greene, L. J. & Truscheit, 48. Magnusson, S., Sottrup-Jensen, L., Petersen, T. E., Morris, H. R. E. (Springer-Verlag, New York), pp. 14-22. & Dell, A. (1974) FEBS Lett. 44, 189-193. 20. Harpel, P. C., Mosesson, M. W. & Cooper, N. R. (1975) in Pro- 49. Stenflo, J., Fernlund, P. & Roepstorff, P. (1975) in Proteases and teases and Biological Control, eds. Reich, E., Rifkin, D. B. & Biological Control, eds. Reich, E., Rifkin, D. B. & Shaw, E. (Cold Shaw, E. (Cold Spring Harbor Laboratory, Cold Spring Harbor, Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), Vol. 2, N.Y.), Vol. 2, pp. 387-404. pp. 111-122. 21. Jeppsson, J. and Laurell, C. (1975) in Proteases and Biological 50. Freer, S. T., Kraut, J., Robertus, J. D., Wright, H. T. & Xuong, Control, eds. Reich, E., Rifkin, D. B. & Shaw, E. (Cold Spring N. H. (1970) Biochemistry 9, 1997-2009. Harbor Laboratory, Cold Spring Harbor, N.Y.), Vol. 2, pp. 51. Robinson, N. C., Neurath, H. & Walsh, K. A. (1973) Biochemistry 405-414. 12,420-426. 22. Huber, R., Kikla; D., Steigemann, W., Deisenhofer, J. & Jones, 52. Morgan, P. H., Robinson, N. C., Walsh, K. A. & Neurath, H. A. in Bayer Symposium V, Proteinase Inhibitors, eds. Fritz, H., (1972) Proc. Natl. Acad. Sci. USA 69,3312-316. Tschesche, H., Green, L. J. & Truscheit, E. (Springer-Verlag, New 53. Kassell, B. & Kay, J. (1973) Science 180, 1022-1027. York), pp. 497-512. 54. Gertler, A., Walsh, K. A. & Neurath, H. (1974) Biochemistry 13, 23. Sweet, R. M., Wright, H. T., Janin, J., Chothia, C. H. & Blow, D. 1302-1310. M. (1974) Biochemistry 13,4212-4228. 55. Kerr, M. A., Walsh, K. A. & Neurath, H. (1975) Biochemistry 14, 24. Hass, G. M., Ako, H., Grahn, D. T. & Neurath, H. (1976) Bio- 5088-5094. chemistry 15,93-100. 56. Morgan, P. H., Walsh, K. A. & Neurath, H. (1974) FEBS Lett. 25. Nagasawa, S., Takahashi, H., Koida, M., Suzuki, T. & Schoen- 41, 108-110. makers, J. G. G. (1968) Biochem. Biophys. Res. Commun. 32, 57. Henderson, R., Wright, C. S., Hess, G. P. & Blow, D. M. (1972) 644-649. Cold Spring Harbor Symp. Quant. Biol. 36,63-70. 26. Ulevitch, R. J., Cochrane, C. G., Revak, S. D., Morrison, D. C. & Johnston, A. R. (1975) in Proteases and Biological Control, eds. 58. Robertus, J. D., Kraut, J., Alden, R. A. & Birktoft, J. J. (1972) Reich, E., Rifkin, D. B. & Shaw, E. (Cold Spring Harbor Labo- Biochemistry 11,4293-4303. ratory, Cold Spring Harbor, N.Y.), Vol. 2, pp. 85-93. 59. Nagendra, K., Reddy, N. & Markus, G. (1972) J. Biol. Chem. 247, 27. Pisano, J. J. (1975) in Proteases and Biological Control, eds. 1683-1691. Reich, E., Rifkin, D. B. & Shaw, E. (Cold Spring Harbor Labo- 60. Summaria, L., Arzadon, L., Bernabe, P. & Robbins, K. C. (1974) ratory, Cold Spring Harbor, N.Y.), Vol. 2, pp. 199-222. J. Biol. Chem. 249,4760-4769. 28. Astrup, T. (1975) in Proteases and Biological Control, eds. Reich, 61. Skeggs, L. T., Jr., Kahn, J. R., Lentz, K. & Shumway, N. P. (1957) E., Rifkin, D. B. & Shaw, E. (Cold Spring Harbor Laboratory, J. Exp. Med. 106,439-453. Cold Spring Harbor, N.Y.), Vol. 2, pp. 343-355. 62. Tager, H. S. & Steiner, D. F. (1973) Proc. Natl. Acad. Sci. USA 29. Dan, K. & Reich, E. (1975) in Proteases and Biological Control, 70,2321-2325. eds. Reich, E., Rifkin, D. B. & Shaw, E. (Cold Spring Harbor 63. Holstein, T. J., Stowell, C. P., Quevedo, W. C., Jr., Zarcaro, R. M. Laboratory, Cold Spring Harbor, N.Y.), Vol. 2, pp. 357-366. & Bienieki, T. C. (1973) Yale J. Biol. Med. 46,560-571. & 64. McGuire, J., Newman, D. & Barisas, G. (1973) Yale J. Biol. Med. 30. Felsted, R. L., Kramer, K. J., Law, J. H., Berger, E. Kafatos, 46,572-582. F. C. (1973) J. Biol. Chem. 248,3012-3020. K. (1973) Arch. Biochem. 31. Meizel, S. & Mukerji, S. K. (1975) Biol. Reprod. 13,83-93. 65. Dohke, Biophys. 157,203-209. 32. Hugh, T. E. (1975) in Proteases and Biological Control, eds. 66. Leckie, B. (1973) Clin. Sci. 44,301-304. Reich, E., Rifkin, D. B. & Shaw, E. (Cold Spring Harbor Labo- 67. de Haas, G. H., Slotboom, A. J., Bonsen, P. P. M., Van Deenen, ratory, Cold Spring Harbor, N.Y.), Vol. 2, pp. 273-290. L. L. M., Maroux, S., Puigserver, A. & Desnuelle, P. (1970) Bio- 33. Jesty, J. & Esnouf, M. P. (1973) Blochem. J. 131, 791-799. chim. Biophys. Acta 221,31-53. 34. Blobel, G. & Dobberstein, B. (1975) J. Cell Biol. 67, 835-862. 68. Harper, E., Bloch, K. J. & Gross, J. (1971) Biochemistry 10, 35. Devillers-Thiery, A., Kindt, T., Scheele, G. & Blobel, G. (1975) 3035-3041. Proc. Natl. Acad. Sci. USA 72,5016-5020. 69. Kreil, G. (1973) Eur. J. Biochem. 33,558-566. 36. Kemper, B., Habener, J. F., Ernst, M. D., Potts, J. T., Jr. & Rich, 70. Dayhoff, M. 0. (1972) Atlas of Protein Sequence and Structure A. (1976) Biochemistry 15, 15-19. (National Biomedical Research Foundation, Washington, D.C.), 37. Schechter, I. & Burstein, Y. (1976) Biochem. Biophys. Res. Vol. 5. Commun. 68, 489-496. 71. Hermodson, M. A., Ericsson, L. H., Neurath, H. & Walsh, K. A. 38. Neurath, H., Walsh, K. A. & Winter, W. P. (1967) Science 158, (1973) Biochemistry 12,3146-3153. 1638-1644. 72. Titani, K., Ericsson, L. H., Neurath, H. & Walsh, K. A. (1975) 39. Hartley, B. S. & Shotton, D. M. (1971) in The Enzymes, ed. Boyer, Biochemistry 14, 1358-1366. P. D. (Academic Press, New York), 3rd ed., Vol. 3, pp. 323- 73. Robbins, K. C., Bernabe, P., Arzadon, L. & Summaria, L. (1972) 373. J. Biol. Chem. 247,6757-6762. 40. Blow, D. M. (1976) Acc. Chem. Res. 9, 145. 74. Robbins, K. C., Bernabe, P., Arzadon, L. & Summaria, L. (1973) 41. Enfield, D. L., Ericsson, L. H., Fujikawa, K., Titani, K., Walsh, J. Biol. Chem. 248,1631-1633. K. A. & Neurath, H. (1974) FEBS Lett. 47, 132-135. 75. Kramer, K. J., Felsted, R. L. & Law, J. H. (1973) J. Biol. Chem. 42. Titani, K., Hermodson, M. A., Fujikawa, K., Ericsson, L. H., 248,3021-3028. Walsh, K. A., Neurath, H. & Davie, E. W. (1972) Biochemistry 76. Wallen, P. & Wiman, B. (1975) in Proteases and Biological 11,4899-4903. Control, eds. Reich, E., Rifkin, D. B. & Shaw, E. (Cold Spring 43. Enfield, D. L., Ericsson, L. H., Walsh, K. A., Neurath, H. & Ti- Harbor Laboratory, Cold Spring Harbor, N.Y.), Vol. 2, pp. tani, K. (1975) Proc. Natl. Acad. Sci. USA 72, 16-19. 291-303. 44. Titani, K., Fujikawa, K., Enfield, D. L., Ericsson, L. H., Walsh, 77. Perlmann, G. E. & Lorand, L., eds. (1970) Methods in Enzy- K. A. & Neurath, H. (1975) Proc. Natl. Acad. Sci. USA 72, mology (Academic Press, New York), Vol. 19. 3082-3086. 78. Boyer, P. D., ed. (1971) The Enzymes, 3rd ed. (Academic Press, 45. Neurath, H., Walsh, K. A. & Gertler, A. (1974) in Metabolic In- New York), Vol. 3. terconversion of Enzymes, 1973, eds. Fischer, E. H., Krebs, E. 79. Reich, E., Rifkin, D. & Shaw, E., eds. (1975) Cold Spring Harbor G., Neurath, H. & Stadtman, E. R. (Springer-Verkg, New York), Conferences on Cell Proliferation (Cold Spring Harbor Labo- pp. 301-312. ratory, Cold Spring Harbor, N.Y.), Vol. 2. Downloaded by guest on September 27, 2021